20 Oct
2010
20 Oct
'10
3:21 p.m.
On Wed, 2010-10-20 at 16:18 +0200, Jordi Delgado wrote:
*** glibc detected *** mri_em_register: malloc(): memory corruption: 0x00000000315dfdc0 *** ======= Backtrace: ========= /lib64/libc.so.6[0x3304072fae] /lib64/libc.so.6(__libc_calloc+0xcd)[0x330407495d] mri_em_register[0x42efad] mri_em_register[0x40e80d] /lib64/libc.so.6(__libc_start_main+0xf4)[0x330401d994] mri_em_register(__gxx_personality_v0+0x16a)[0x40b16a]
<snip>
That's odd. However, mri_em_register is the plain CPU binary - the CUDA binaries have _cuda appended. Do you get the same problem if you don't enable CUDA when running mri_em_register?
Regards,
Richard